Digital twins to study the brain and brain disorders

An international team of neuroscientists proposes a new computational model to more precisely study unknown aspects of the human brain and the causes of various neuropsychiatric disorders. The researchers call on the scientific community to use this new computational model to search for personalized diagnoses and treatments of various disorders (such as coma, epilepsy, Parkinson’s disease or multiple sclerosis, among others), in the line of so-called precision medicine.

The team is made up of specialists from the Pompeu Fabra University (UPF) in Barcelona, ​​the University of Girona (UdG) and the University of Oxford.

These researchers start from an innovative approach in neuroscience, called the whole brain model, which is based on computational simulations of its functioning and allows the creation of virtual brain twins of specific patients. Unlike traditional approaches, this model allows us to analyze brain dynamics as a whole, beyond those of specific regions, and understand how they interact with each other. The creation of this model is the result of work developed over decades by neuroscientists in this group, for whom computational neuroscience is already at a point of sufficient maturity to reveal unknown aspects of the human brain in health and disease.

Recently, the research team has published details of how this model is being applied, for example to predict how to wake a patient in a coma by stimulating specific parts of his brain, or to understand how seizures spread through the brains of subjects with epilepsy. and try ways to stop them.

The model is based on creating a computational simulation of the human brain with neuroimaging data (mainly from MRIs) and mathematical formulas. The result is a virtual twin of a specific patient’s brain, showing a detailed map of not only their anatomy, but also their activity. Making a simile with road traffic, we could consider that the map not only shows the roads but also the traffic.

The authors of the study propose to study, using digital twins of patients, their brain and various brain disorders. (Illustration: Amazings/NCYT)

The model allows simulations to be carried out to analyze how different brain regions communicate and interact with each other or to test them in different scenarios, for example to analyze how a person’s brain would respond to certain changes or stimuli. These tests are performed in highly controlled virtual environments and allow the brain reactions of a specific patient to be explored with personalized, safe and non-invasive techniques.

The model allows us to analyze with great precision both brain activity at a specific moment and its fluctuations over time on very small time scales of milliseconds. This makes it particularly suitable for examining the process of transition between brain states, for example between sleeping and awake, or between coma and full consciousness. In fact, some tests carried out by the research team have been based on neuroimages of sleeping people. Based on these data, it has been explored, through such computational simulations, which are the specific parts of the brain that need to be stimulated to cause coma patients to wake up. Other things have also been explored.

Gustavo Deco, from the research team and director of the Computational Neuroscience group at the Center for Brain and Cognition (CBC) at UPF, explains: “Computational neuroscience has reached a point of maturity in which it can provide tools that finally reach to unravel fundamental principles of brain functions in health and disease.”

Gustavo Patow, from the research team and member of the Visualization, Virtual Reality and Graphic Interaction group (ViRVIG) of the UdG and the CBC of the UPF, adds: “One of the great challenges when it comes to understanding brain diseases, like depression or epilepsy, is that we don’t fully understand what happens inside the brain. Traditional methods have struggled to offer clear answers. “Whole-brain models offer a new way of looking at the brain, helping researchers identify what might be wrong and how to fix it.”

Researchers have already demonstrated the promising potential of the whole-brain model to treat several neuropsychiatric disorders so far without a cure. As it is refined, the model could become a standard tool in hospitals and help professionals better diagnose and treat brain diseases.

Deco, Patow and their colleagues present the technical details of their latest progress and what it means in the academic journal Nature Reviews Methods Primers, under the title “Whole-brain modeling: an essential tool for understanding brain dynamics.” (Source: Pompeu Fabra University)